Injecting dection for Nvtacker

Accelerated Computing Intelligent Video Analytics DeepStream SDK
,
1

Hi,

I’m creating a pipeline (C/C++) that does not use nvinfer but inject detections from fusion detector and would like to use the Nvtracker for tracking objects:

If I use UNTRACKED_OBJECT_ID the object are not used by the tracker and nothing appears on the src pad, BUT if I set the ObjectID to 1 for instance the objects are not rejected but not really tracked. but the nvOSD is displaying the BB.

?? any idea ? topic already find in the forum but we no definite answers …

Jetson Orin Nano
CUDA: 12.6.68
TensorRT: 10.2.0.30

Using gst-nvtracker for object tracking with config file: nvtracker_config.yml
input source /dev/video0
Creating GStreamer capture for device: /dev/video0
gst_capture_create: device=/dev/video0, size=1280x720, fps=30
DEBUG: Pipeline created: 0xaaaad2422200
Creating pipeline: v4l2src device=/dev/video0 io-mode=4 ! image/jpeg,width=1280,height=720,framerate=30/1 ! jpegdec ! nvvidconv ! video/x-raw(memory:NVMM),format=RGBA ! tee name=t t. ! queue ! appsink name=sink t. ! queue name=mux_queue nvstreammux name=stream-muxer width=1280 height=720 batch-size=1 batched-push-timeout=33333 live-source=1 ! nvtracker name=tracker ! nvdsosd name=osd display-bbox=true display-clock=false process-mode=0 ! nv3dsink sync=false
Configuring appsink...
Configuring nvtracker with library and config file...
nvtracker configured with NvMultiObjectTracker library
Starting pipeline...
gstnvtracker: Loading low-level lib at /opt/nvidia/deepstream/deepstream/lib/libnvds_nvmultiobjecttracker.so
[NvMultiObjectTracker] Initialized

…

DEBUG: Set frame_meta->num_obj_meta = 10 for frame 4
[nvtracker probe] Frame 14 has 10 obj_meta(s)
  -> Object ID: 18446744073709551615, class_id: 57, component_id: 1, conf: 0.56, rect=(116.0,393.0,420.0,185.0)
  -> Object ID: 18446744073709551615, class_id: 57, component_id: 1, conf: 0.58, rect=(1026.0,348.0,252.0,226.0)
  -> Object ID: 18446744073709551615, class_id: 57, component_id: 1, conf: 0.64, rect=(664.0,388.0,342.0,96.0)
  -> Object ID: 18446744073709551615, class_id: 58, component_id: 1, conf: 0.60, rect=(1126.0,308.0,116.0,150.0)
  -> Object ID: 18446744073709551615, class_id: 56, component_id: 1, conf: 0.66, rect=(1024.0,347.0,254.0,227.0)
  -> Object ID: 18446744073709551615, class_id: 56, component_id: 1, conf: 0.80, rect=(316.0,349.0,68.0,54.0)
  -> Object ID: 18446744073709551615, class_id: 56, component_id: 1, conf: 0.88, rect=(208.0,349.0,68.0,52.0)
  -> Object ID: 18446744073709551615, class_id: 56, component_id: 1, conf: 0.83, rect=(0.0,336.0,44.0,222.0)
  -> Object ID: 18446744073709551615, class_id: 39, component_id: 1, conf: 0.54, rect=(728.0,516.0,34.0,59.0)
  -> Object ID: 18446744073709551615, class_id: 39, component_id: 1, conf: 0.73, rect=(764.0,379.0,36.0,50.0)

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU)
• DeepStream Version
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs)
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)

3

Please compare the NvDsBatchMeta which added by you and the output of nvinfer. Please ensure set: frame_meta->bInferDone = TRUE;

4

Thanks that is working now setting binferDone !!!